Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
On Tue, 2004-03-23 at 20:10, Eugene Crosser wrote: > > > Please could someone knowledgeble tell me why this may happen? > > > Under load (local massive quota update by a script plus cpio of a big > > > tree from an NFS client) synchronization is repeatedly restarted after > > > ~1% completion: I have some follow-up information: I figured that it was not actually quota editing that triggered the problem, it was "sync" command issued on every 1000th update in the script. After the system actively works for about an hour (with quota updating or just regular file operations), "sync" begins to take several minutes to complete, "top" shows that it uses 99% cpu and system response becomes very very slow. So maybe(?) this is not drbd problem per se, but a sync() run on actively updated ext3 filesystem (with quota). OTOH it may be because sync creates a load pattern that drbd device is not prepared to... Eugene -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20040323/653fe5b8/attachment.pgp>